Natural disaster risk monitoring system based on satellite remote sensing image

Abstract

The invention discloses a natural disaster risk monitoring system based on a satellite remote sensing image, and relates to the technical field of natural disaster monitoring, and the system comprises a data analysis module which obtains environment data according to a satellite remote sensing platform and a sensor platform, and comprehensively analyzes soil humidity; and the drought risk prediction module is used for analyzing the availability of water resources on the basis of monitoring the water demand of the agricultural region. According to the method, basic data support is provided for climate change analysis through multispectral remote sensing data and sensor platform data, the water resource degree in the current detection area can be visually judged through comprehensive analysis of water resource data, and the water resource degree in the current detection area can be predicted through a time sequence prediction model. The drought risk can be early warned in advance by combining comprehensive water resource availability, multispectral remote sensing data and sensor platform data, cloud layer data are analyzed through remote sensing data, and whether the monitored cloud layer environment of the agricultural area is suitable for artificial rainfall or not is judged by combining the spatial relationship between the cloud layer and the agricultural area.

Description

Technical Field

[0001] The present invention relates to the technical field of natural disaster monitoring, in particular to a natural disaster risk monitoring system based on satellite remote sensing images. Background Art

[0002] With global climate change and population growth, the demand for agricultural production and environmental protection continues to increase. Modern agricultural management has gradually relied on advanced technologies to cope with the risks of natural disasters. Satellite remote sensing technology, as an efficient and wide-area data acquisition method, has been widely used in agricultural monitoring. Combined with field environmental data obtained by sensor platforms, it can further improve the accuracy and timeliness of data, thereby building a more comprehensive environmental monitoring system.

[0003] Among them, the natural disaster risks in agricultural areas are mostly composed of drought risks. In order to cope with drought risks, emergency measures such as surrounding water resources and artificial rainfall are usually considered. However, when monitoring the risks of agricultural natural disasters, existing agricultural natural disaster monitoring technologies usually only focus on disaster risks in related fields. After the drought risk occurs, it is difficult to timely monitor the emergency measures for drought risks, resulting in a waste of satellite remote sensing resources. It is also impossible to provide dynamic and real-time monitoring information on emergency measures for disaster risks in agricultural areas. Summary of the Invention

[0004] In view of the above-mentioned problems existing in the existing natural disaster risk monitoring system based on satellite remote sensing images, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the existing agricultural natural disaster monitoring technology usually only focuses on the disaster risks in related fields when responding to the risk monitoring of agricultural natural disasters. After the drought risk occurs, it is difficult to timely monitor the emergency measures conditions for the drought risk, resulting in a waste of satellite remote sensing resources, and it is impossible to dynamically and real-time provide relevant monitoring information on emergency measures for disaster risks in agricultural areas, thereby reducing the monitoring efficiency.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a natural disaster risk monitoring system based on satellite remote sensing images, comprising:

[0007] The data analysis module obtains environmental data from satellite remote sensing platforms and sensor platforms and comprehensively analyzes soil moisture;

[0008] The drought risk prediction module monitors water demand in agricultural areas, analyzes water resource availability, and constructs a long-short-term memory network model to predict and monitor drought risks in agricultural areas.

[0009] The artificial rainfall suitability analysis module uses a satellite remote sensing platform to identify cloud areas, analyze cloud parameters, identify clouds suitable for artificial rainfall, build a suitability prediction model to predict the suitability of clouds for artificial rainfall, analyze the spatial overlap of clouds and monitored agricultural areas, and determine the feasibility of artificial rainfall;

[0010] The early warning information sending module issues early warning information based on water resource availability, drought risk and artificial rainfall feasibility, and generates templates based on the early warning information and sends them to management personnel.

[0011] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, wherein: the environmental data is obtained according to the satellite remote sensing platform and the sensor platform, and the soil moisture is comprehensively analyzed, including:

[0012] Acquire multispectral remote sensing image data for monitoring agricultural areas based on satellite remote sensing platforms, including visible light, infrared, and microwave data;

[0013] Preprocess multispectral remote sensing image data, including radiation correction, atmospheric correction and geometric correction;

[0014] Based on the satellite remote sensing platform, the reflectance of the near-infrared band, red light band and green light band is obtained, and the vegetation index NDVI and water body index NDWI are calculated;

[0015] Based on the sensor platform, the soil moisture data of different layers in the agricultural area are collected and analyzed comprehensively, which is expressed as:

[0016]

[0017] Among them SM a represents the integrated soil moisture, SM i represents the moisture content of the i-th soil layer, and n represents the total number of layers.

[0018] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, wherein: the water demand of the agricultural area is monitored and the water resource availability is analyzed, including:

[0019] Based on the sensor platform, available water resource data of monitored agricultural areas is obtained, including river water resource data, lake water resource data, and reservoir water resource data, which can be expressed as:

[0020] WR r =WL·F;

[0021] WR l =LL·A;

[0022] WR t =RL·V;

[0023] Among them WR r Indicates the water availability of the river, WR l Represents the water availability of the lake, WR t represents the water availability of the reservoir, WL represents the river level, LL represents the lake level, RL represents the reservoir level, F represents the water flow, A represents the lake area, and V represents the reservoir capacity;

[0024] The comprehensive water availability is analyzed based on the available water resources in the monitored agricultural areas and expressed as:

[0025] WR a =WR r +WR l +WR t ;

[0026] Among them WR a represents the comprehensive water resources availability;

[0027] Based on the historical data of available water resources, the mean and standard deviation of the historical comprehensive water resources availability data are calculated, and the comprehensive water resources availability data are standardized to calculate WR' a ;

[0028] Taking into account the historical comprehensive water resources availability data and agricultural water use data, a historical threshold is set, which is expressed as:

[0029]

[0030] Where T represents the historical threshold, m represents the total number of historical time periods, represents the comprehensive water resource availability in time period i, AW i represents the agricultural water consumption in the i-th period, AW w represents the historical average agricultural water use;

[0031] The threshold is adjusted according to the crop water demand in the currently monitored agricultural area and expressed as:

[0032]

[0033] Where T a Indicates real-time threshold, AW c Indicates the amount of agricultural water used in the current time period;

[0034] If WR' a Greater than T a , it means that the available water resources in the monitored agricultural areas can meet the agricultural water demand;

[0035] If WR' a Less than or equal to T a, it means that the available water resources in the monitored agricultural area cannot meet the agricultural water demand.

[0036] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, the long-short term memory network model is constructed to predict and monitor the drought risk in agricultural areas, including:

[0037] Comprehensive water resources availability data WR' based on standardized processing a , construct a long short-term memory network model to predict and monitor soil moisture in agricultural areas, including input layer, LSTM layer, fully connected layer and output layer;

[0038] The standardized comprehensive water resources availability data WR' a , vegetation index NDVI, water body index NDWI, comprehensive soil moisture SM a The land surface temperature data collected by sensors monitoring agricultural areas are time-aligned and fed into the model through the input layer;

[0039] Use the training set to train the long short-term memory network model using the mean squared error loss function;

[0040] Newly collected data are fed into the model to predict soil moisture in monitored agricultural areas;

[0041] Based on historical data, the mean value of soil moisture minus the standard deviation is used as the high-risk threshold, and the mean value of soil moisture is used as the medium-risk threshold;

[0042] If the predicted value of soil moisture y i If the value is greater than or equal to the high-risk threshold, the monitored agricultural area is judged to be at high risk of drought;

[0043] If the predicted value of soil moisture y i If the drought risk is greater than or equal to the medium risk threshold and less than the high risk threshold, the monitored agricultural area is judged to be at medium drought risk;

[0044] If the predicted value of soil moisture y i If the drought risk is less than the medium risk threshold, the monitored agricultural area is judged to be at low risk of drought.

[0045] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, wherein: the identification of cloud areas through the satellite remote sensing platform includes:

[0046] Based on the multispectral remote sensing image data after radiation correction and atmospheric correction, the identification bands are selected for analysis. The identification bands include visible light band, near infrared band and infrared band.

[0047] Set corresponding thresholds for visible light band, near infrared band and infrared band based on historical data;

[0048] The identified band data that is greater than the corresponding threshold is binarized to segment and identify the cloud area.

[0049] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, the analysis of cloud parameters and identification of clouds suitable for artificial rainfall includes:

[0050] Based on the satellite remote sensing platform, infrared band data is used to obtain cloud top temperature, and the cloud top height is calculated by combining the atmospheric temperature profile data, which is expressed as:

[0051]

[0052] Where CTH is the cloud top height, H0 is the tropopause height, T0 is the tropopause temperature, LR is the temperature lapse rate, and CTT is the cloud top temperature;

[0053] The cloud base height is calculated using the wet bulb temperature method and is expressed as:

[0054]

[0055] Where CBH is the cloud base height, T is the surface temperature, and T d Indicates dew point temperature, F indicates temperature lapse rate;

[0056] Based on the satellite remote sensing platform, microwave remote sensing data is used to obtain the water vapor density of each altitude layer in the cloud layer, and the cloud water path is calculated, which is expressed as:

[0057]

[0058] Where CWP represents cloud water path, p c (z) represents the water vapor density at the cloud height z;

[0059] Based on the artificial rainfall operation conditions, the cloud water path, cloud thickness and cloud top temperature thresholds are set respectively, and the clouds that meet the threshold judgment conditions are marked as clouds suitable for artificial rainfall.

[0060] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, wherein: the construction of the suitability prediction model to predict the suitability of artificial rainfall in the cloud layer includes:

[0061] A suitability prediction model is constructed to judge the suitability of artificial rainfall based on the artificial rainfall operation conditions, which is expressed as:

[0062] TR=w1·CWP+w2·CT+w3·CTT;

[0063] Where TR represents the suitability of artificial rainfall, CT is the difference between the cloud top height and the cloud base height, and w1, w2 and w3 are weight parameters;

[0064] Use the training set to train the model by minimizing the objective function, expressed as:

[0065]

[0066] Where c represents the total number of samples, r i Indicates the actual value, Represents the model output;

[0067] The model parameters are iteratively optimized using the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the fitness prediction model.

[0068] The suitability of artificial rainfall was calculated using the suitability prediction model based on historical sample data, and a high suitability threshold T was set based on the 90th percentile. a , set the suitability threshold T based on the sample median z ;

[0069] If the artificial rainfall suitability TR is greater than or equal to the high suitability threshold, the monitored cloud layer is determined to be suitable for artificial rainfall;

[0070] If the artificial rainfall suitability TR is greater than or equal to the medium suitability threshold and less than the high suitability threshold, it is determined that artificial rainfall can be carried out in the monitored cloud layer;

[0071] If the artificial rainfall suitability TR is less than the medium suitability threshold, the monitored cloud layer is judged to be unsuitable for artificial rainfall.

[0072] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, the analysis of the spatial superposition of cloud layers and monitored agricultural areas to determine the feasibility of artificial rainfall includes:

[0073] Obtain cloud position coordinates from multispectral remote sensing image data based on the identified cloud area;

[0074] Based on the GIS boundary data of the monitored agricultural area, the cloud position coordinates are mapped into the GIS system and displayed on the same map with the GIS boundary data of the monitored agricultural area. Spatial overlay analysis is performed, and the spatial overlay with the cloud data is judged according to the monitored agricultural area, including inclusion judgment and intersection judgment;

[0075] The median of the intersection judgment is set as the boundary threshold. If the intersection judgment value of the cloud data and the monitored agricultural area is greater than the boundary threshold, and the intersection judgment of the cloud data and the monitored agricultural area is included, the analysis result is suitable for artificial rainfall.

[0076] If the intersection of cloud data and the monitored agricultural area is less than or equal to the boundary threshold, the analysis result is that it is not suitable for artificial rainfall.

[0077] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images of the present invention, the early warning information based on water resource availability, drought risk and artificial rainfall feasibility includes:

[0078] Based on the comprehensive water resource availability and real-time threshold determination of the monitored agricultural area, an agricultural water shortage warning message is issued when the available water resources in the monitored area cannot meet the agricultural water demand;

[0079] Based on the predicted soil moisture values ​​of monitored agricultural areas and the determination of high and low risk thresholds, drought warning information will be issued when it is determined to be a moderate or high drought risk;

[0080] Based on the cloud areas identified in the monitoring of agricultural areas, and the determination of the suitability of the cloud layers for artificial rainfall and the high suitability threshold and the medium suitability threshold, when it is determined that artificial rainfall is not suitable, an early warning message that artificial rainfall is not suitable is issued.

[0081] As a preferred solution of the natural disaster risk monitoring system based on satellite remote sensing images described in the present invention, the above-mentioned generating a template based on the warning information and sending it to the management personnel refers to predefining the warning information template through the Python database based on the issued warning information, inputting the warning information content into the warning information template, configuring the SMTP server, sending the warning information template with the completed content input via email to the management personnel's email account, and sending the warning information template with the completed content input via SMS to the management personnel's mobile phone through SMS sending service.

[0082] The beneficial effects of the present invention are as follows: basic data support is provided for climate change analysis through multispectral remote sensing data and sensor platform data; the water resource level in the current detection area can be intuitively judged through comprehensive analysis of water resource data; drought risk can be warned in advance through a time series prediction model combined with comprehensive water resource availability, multispectral remote sensing data and sensor platform data; cloud data is analyzed through remote sensing data, and the spatial relationship between clouds and agricultural areas is combined to judge whether the cloud environment in the monitored agricultural area is suitable for artificial rainfall, so as to achieve the effect of monitoring drought risk in the monitored area while monitoring the natural conditions for artificial rainfall, thereby making full use of the remote sensing data of the satellite remote sensing platform, improving data utilization and the efficiency of monitoring the emergency measures environment in the monitored area. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0084] Figure 1 Schematic diagram of the structure of the natural disaster risk monitoring system based on satellite remote sensing images.

[0085] Figure 2 This is a flow chart of the natural disaster risk monitoring system based on satellite remote sensing images. DETAILED DESCRIPTION

[0086] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0087] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0088] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it constitute a separate or selective embodiment that is mutually exclusive with other embodiments.

[0089] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a natural disaster risk monitoring system based on satellite remote sensing images. The natural disaster risk monitoring system based on satellite remote sensing images includes:

[0090] S1, obtains environmental data from satellite remote sensing platforms and sensor platforms and comprehensively analyzes soil moisture;

[0091] Preferably, environmental data is obtained from satellite remote sensing platforms and sensor platforms, and soil moisture is comprehensively analyzed, including:

[0092] Acquire multispectral remote sensing image data for monitoring agricultural areas based on satellite remote sensing platforms, including visible light, infrared, and microwave data;

[0093] Preprocess multispectral remote sensing image data, including radiation correction, atmospheric correction and geometric correction;

[0094] Based on the satellite remote sensing platform, the reflectance of the near-infrared band, red light band and green light band is obtained, and the vegetation index NDVI and water body index NDWI are calculated;

[0095] Based on the sensor platform, the soil moisture data of different layers in the agricultural area are collected and analyzed comprehensively, which is expressed as:

[0096]

[0097] Among them SM a represents the integrated soil moisture, SM i represents the moisture content of the i-th soil layer, and n represents the total number of layers.

[0098] By comprehensively analyzing soil moisture and vegetation index, soil drought conditions can be discovered in a timely manner and early warnings can be issued, helping the agricultural sector to take effective measures to reduce the impact of drought on agricultural production. By monitoring the water body index NDWI, the distribution and changes of water resources around farmland can be understood, helping the agricultural sector to rationally manage and protect water resources and prevent water waste and pollution. Using multispectral remote sensing image data and soil moisture data, farmland can be comprehensively planned and optimized. Through multispectral remote sensing data and soil moisture data, a large amount of data can be accumulated to provide basic data support for climate change analysis and help analyze the impact of climate change on agricultural production.

[0099] S2, based on the water demand of monitored agricultural areas, analyzes water resource availability, constructs a long-short-term memory network model, and predicts drought risks in monitored agricultural areas;

[0100] Preferably, the water availability is analyzed based on monitoring water demand in agricultural areas, including:

[0101] Based on the sensor platform, available water resource data of monitored agricultural areas is obtained, including river water resource data, lake water resource data, and reservoir water resource data, which can be expressed as:

[0102] WR r =WL·F;

[0103] WR l =LL·A;

[0104] WR t =RL·V;

[0105] Among them WR r Indicates the water availability of the river, WR l Represents the water availability of the lake, WR trepresents the water availability of the reservoir, WL represents the river level, LL represents the lake level, RL represents the reservoir level, F represents the water flow, A represents the lake area, and V represents the reservoir capacity;

[0106] The comprehensive water availability is analyzed based on the available water resources in the monitored agricultural areas and expressed as:

[0107] WR a =WR r +WR l +WR t ;

[0108] Among them WR a represents the comprehensive water resources availability;

[0109] Based on the historical data of available water resources, the mean and standard deviation of the historical comprehensive water resources availability data are calculated, and the comprehensive water resources availability data are standardized to calculate WR' a ;

[0110] Taking into account the historical comprehensive water resources availability data and agricultural water use data, a historical threshold is set, which is expressed as:

[0111]

[0112] Where T represents the historical threshold, m represents the total number of historical time periods, represents the comprehensive water resource availability in time period i, AW i represents the agricultural water consumption in the i-th period, AW w represents the historical average agricultural water use;

[0113] The threshold is adjusted according to the crop water demand in the currently monitored agricultural area and expressed as:

[0114]

[0115] Where T a Indicates real-time threshold, AW c Indicates the amount of agricultural water used in the current time period;

[0116] If WR' a Greater than T a , it means that the available water resources in the monitored agricultural areas can meet the agricultural water demand;

[0117] If WR' a Less than or equal to T a , it means that the available water resources in the monitored agricultural area cannot meet the agricultural water demand.

[0118] By comprehensively analyzing water resource data including rivers, lakes and reservoirs, the types of available water resources in the monitored agricultural area are considered more comprehensively. By comprehensively considering the availability of water resources and the setting of corresponding thresholds, a judgment standard is provided for the available water resources in the monitoring area. According to the comprehensive water resource availability calculation combined with the threshold, the water resource level in the current detection area can be intuitively judged, and the water resource management strategy can be adjusted in time based on the judgment of water resource availability to ensure the stability and sustainability of agricultural production. Through real-time monitoring and historical data analysis, farmers and agricultural departments can be helped to understand whether water resources are sufficient in a timely manner, and take preventive measures to reduce the impact of water shortage on agricultural production and improve the scientificity and effectiveness of agricultural management.

[0119] Furthermore, a long-short-term memory network model was constructed to predict and monitor drought risks in agricultural areas, including:

[0120] Comprehensive water resources availability data WR' based on standardized processing a , construct a long short-term memory network model to predict and monitor soil moisture in agricultural areas, including input layer, LSTM layer, fully connected layer and output layer;

[0121] The standardized comprehensive water resources availability data WR' a , vegetation index NDVI, water body index NDWI, comprehensive soil moisture SM a The land surface temperature data collected by sensors monitoring agricultural areas are time-aligned and fed into the model through the input layer;

[0122] The LSTM layer is used to process time series data and capture the temporal dependencies between features. The fully connected layer performs nonlinear transformation on the output data of the LSTM layer to generate intermediate feature vectors. The output layer converts the output feature vectors of the fully connected layer into soil moisture prediction values.

[0123] The long short-term memory network model is trained using the training set through the mean square error loss function, which is expressed as:

[0124]

[0125] Where MSR represents the calculation error, v represents the total number of samples, and y i Indicates the actual value, represents the predicted value;

[0126] The Adam optimizer and the gradient descent method are used to iteratively optimize the model parameters. When the loss of the LSTM model no longer decreases significantly during the continuous iteration, the iteration is stopped and the model parameters are output to update the LSTM model.

[0127] Newly collected data are fed into the model to predict soil moisture in monitored agricultural areas;

[0128] Based on historical data, the mean value of soil moisture minus the standard deviation is used as the high-risk threshold, and the mean value of soil moisture is used as the medium-risk threshold;

[0129] If the predicted value of soil moisture y i If the value is greater than or equal to the high-risk threshold, the monitored agricultural area is judged to be at high risk of drought;

[0130] If the predicted value of soil moisture y i If the drought risk is greater than or equal to the medium risk threshold and less than the high risk threshold, the monitored agricultural area is judged to be at medium drought risk;

[0131] If the predicted value of soil moisture y i If the drought risk is less than the medium risk threshold, the monitored agricultural area is judged to be at low risk of drought.

[0132] The LSTM model's time series prediction model, combined with comprehensive water resource availability, NDVI, NDWI and other data, can provide early warning of drought risks, helping farmers and agricultural departments take timely measures to reduce the impact of drought on agricultural production. The risk threshold determined based on historical data is used to determine the risk of the model's prediction results, which can provide intuitive prediction information for agricultural-related departments and farmers, improve the level of intelligent agricultural management, and optimize irrigation plans and agricultural production strategies by accurately predicting soil moisture and drought risks, thereby increasing crop yield and quality, assisting agricultural managers in making reasonable decisions, and improving the scientific nature and effectiveness of agricultural management.

[0133] S3 uses satellite remote sensing platforms to identify cloud areas, analyze cloud parameters, identify clouds suitable for artificial rainfall, build a suitability prediction model to predict the suitability of clouds for artificial rainfall, analyze the spatial overlap between clouds and monitored agricultural areas, and determine the feasibility of artificial rainfall;

[0134] Preferably, the cloud area is identified by a satellite remote sensing platform, including:

[0135] Based on the multispectral remote sensing image data after radiation correction and atmospheric correction, the identification bands are selected for analysis. The identification bands include visible light band, near infrared band and infrared band.

[0136] Set corresponding thresholds for visible light band, near infrared band and infrared band based on historical data;

[0137] The identified band data that is greater than the corresponding threshold is binarized to segment and identify the cloud area.

[0138] By combining multispectral remote sensing image data after radiation correction and atmospheric correction with thresholds set by historical data, cloud areas can be accurately identified and the accuracy of cloud recognition can be improved. Accurate cloud recognition is of great significance for weather forecasting, climate research, and artificial rainfall, and provides a basis for cloud recognition in artificial rainfall.

[0139] Furthermore, cloud parameters are analyzed to identify clouds suitable for artificial rainfall, including:

[0140] Based on the satellite remote sensing platform, infrared band data is used to obtain cloud top temperature, and the cloud top height is calculated by combining the atmospheric temperature profile data, which is expressed as:

[0141]

[0142] Where CTH is the cloud top height, H0 is the tropopause height, T0 is the tropopause temperature, LR is the temperature lapse rate, and CTT is the cloud top temperature;

[0143] The cloud base height is calculated using the wet bulb temperature method and is expressed as:

[0144]

[0145] Where CBH is the cloud base height, T is the surface temperature, and T d represents the dew point temperature, F represents the temperature lapse rate, and the empirical constant used is 0.0065;

[0146] Based on the satellite remote sensing platform, microwave remote sensing data is used to obtain the water vapor density of each altitude layer in the cloud layer, and the cloud water path is calculated, which is expressed as:

[0147]

[0148] Where CWP represents cloud water path, p c (z) represents the water vapor density at the cloud height z;

[0149] Based on the artificial rainfall operation conditions, the cloud water path, cloud thickness and cloud top temperature thresholds are set respectively, and the clouds that meet the threshold judgment conditions are marked as clouds suitable for artificial rainfall.

[0150] By accurately calculating the cloud top height, cloud base height and cloud water path, and based on the requirements given in the artificial rainfall operation conditions regulations, clouds suitable for artificial rainfall can be effectively identified, and the success rate and efficiency of artificial rainfall operations can be improved. By using multi-source remote sensing data and precise calculation methods, more accurate cloud characteristics analysis can be provided, providing reliable data support for weather forecasts. By accurately identifying cloud height and water vapor content, the accuracy of rainfall forecasts can be improved, helping agricultural and water conservancy departments to prepare response measures, and thus comprehensively judge whether the clouds are suitable for artificial rainfall. Monitoring based on multi-source remote sensing data can help agricultural departments and farmers identify clouds suitable for artificial rainfall, reasonably arrange artificial rainfall operations, and improve resource utilization efficiency. At the same time, it makes full use of remote sensing data from satellite remote sensing platforms, improves data utilization and the efficiency of monitoring the emergency measures environment in the monitored area.

[0151] Furthermore, a suitability prediction model is constructed to predict the suitability of cloud layers for artificial rainfall, including:

[0152] A suitability prediction model is constructed to judge the suitability of artificial rainfall based on the artificial rainfall operation conditions, which is expressed as:

[0153] TR=w1·CWP+w2·CT+w3·CTT;

[0154] Where TR represents the suitability of artificial rainfall, CT is the difference between the cloud top height and the cloud base height, and w1, w2 and w3 are weight parameters;

[0155] Use the training set to train the model by minimizing the objective function, expressed as:

[0156]

[0157] Where c represents the total number of samples, r i Indicates the actual value, Represents the model output;

[0158] The model parameters are iteratively optimized using the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the fitness prediction model.

[0159] The suitability of artificial rainfall was calculated using the suitability prediction model based on historical sample data, and a high suitability threshold T was set based on the 90th percentile. a , set the suitability threshold T based on the sample median z ;

[0160] If the artificial rainfall suitability TR is greater than or equal to the high suitability threshold, the monitored cloud layer is determined to be suitable for artificial rainfall;

[0161] If the artificial rainfall suitability TR is greater than or equal to the medium suitability threshold and less than the high suitability threshold, it is determined that artificial rainfall can be carried out in the monitored cloud layer;

[0162] If the artificial rainfall suitability TR is less than the medium suitability threshold, the monitored cloud layer is judged to be unsuitable for artificial rainfall.

[0163] By comprehensively considering multiple factors such as cloud water path, cloud thickness and cloud top temperature, the suitability prediction model can accurately assess whether the cloud layer is suitable for artificial rainfall, thereby improving the success rate and effectiveness of artificial rainfall operations. Based on the suitability judgment of the model, the resource allocation of artificial rainfall operations can be optimized to ensure that artificial rainfall is carried out at the most appropriate time and place, improve resource utilization efficiency, effectively reduce misjudgments, ensure that only suitable clouds are used for artificial rainfall, avoid waste of resources, help decision makers reasonably arrange artificial rainfall operations, save manpower and material costs, improve the intelligence level of meteorological services, and provide precise services for agricultural production and water resources management.

[0164] Furthermore, the spatial overlap between cloud cover and monitored agricultural areas was analyzed to determine the feasibility of artificial rainfall, including:

[0165] Obtain cloud position coordinates from multispectral remote sensing image data based on the identified cloud area;

[0166] Based on the GIS boundary data of the monitored agricultural area, including vector data or raster data, the cloud position coordinates are mapped into the GIS system. The cloud data can be displayed as polygons (cloud boundaries) and displayed on the same map with the GIS boundary data of the monitored agricultural area. Spatial overlay analysis is performed based on the spatial overlay of the cloud data with the monitored agricultural area judgment, including inclusion judgment, where the cloud polygon is completely within the agricultural area polygon and intersection judgment, where the cloud polygon partially overlaps with the agricultural area polygon;

[0167] The median of the intersection judgment is set as the boundary threshold. If the intersection judgment value of the cloud data and the monitored agricultural area is greater than the boundary threshold, and the intersection judgment of the cloud data and the monitored agricultural area is included, the analysis result is suitable for artificial rainfall.

[0168] If the intersection of cloud data and the monitored agricultural area is less than or equal to the boundary threshold, the analysis result is that it is not suitable for artificial rainfall.

[0169] By accurately identifying and analyzing the spatial relationship between clouds and agricultural areas, the scientific nature and accuracy of artificial rainfall decisions can be improved, ensuring that artificial rainfall is carried out in suitable areas and improving rainfall effects. Combined with multiple factors such as the cloud water path, cloud thickness, and cloud top temperature of the cloud itself, different dimensions of analysis and judgment are made, from whether the cloud has the potential for artificial rainfall to whether the cloud is spatially suitable for monitoring artificial rainfall in agricultural areas. This ensures that in the case of water resource shortage, artificial rainfall is given priority in areas with severe drought and insufficient water resources to ensure the sustainability of agricultural production. In addition, natural information related to clouds is monitored simultaneously when natural disasters occur in agricultural areas. Remote sensing data is fully utilized to provide relevant data in a timely manner to solve natural disasters such as drought, provide sufficient information for responding to natural disasters, and assist relevant personnel in timely judging solutions to artificial rainfall.

[0170] S4, issues early warning information based on water resource availability, drought risk, and artificial rainfall feasibility, and generates templates based on the early warning information and sends them to management personnel;

[0171] Preferably, early warning information is issued based on water availability, drought risk and the feasibility of artificial rainfall.

[0172] include,

[0173] Based on the comprehensive water resource availability and real-time threshold determination of the monitored agricultural area, an agricultural water shortage warning message is issued when the available water resources in the monitored area cannot meet the agricultural water demand;

[0174] Based on the predicted soil moisture values ​​of monitored agricultural areas and the determination of high and low risk thresholds, drought warning information will be issued when it is determined to be a moderate or high drought risk;

[0175] Based on the cloud areas identified in the monitoring of agricultural areas, and the determination of the suitability of the cloud layers for artificial rainfall and the high suitability threshold and the medium suitability threshold, when it is determined that artificial rainfall is not suitable, an early warning message that artificial rainfall is not suitable is issued.

[0176] Through multi-level and multi-dimensional data analysis and judgment, comprehensive and accurate early warning information is provided to help agricultural management departments and farmers take timely response measures, thereby improving the coverage and reliability of the early warning system and ensuring that accurate early warning information can be provided under various meteorological and agricultural conditions. By issuing timely early warnings on insufficient agricultural water use, farmers and agricultural management departments are helped to reasonably arrange water use plans. Through drought early warnings, irrigation and drought resistance measures are taken in advance to reduce the impact of drought on agricultural production and increase crop yield and quality. Through early warnings on unsuitable artificial rainfall, artificial rainfall operations can be avoided under unsuitable meteorological conditions, thereby improving the success rate and effectiveness of artificial rainfall.

[0177] Furthermore, generating a template based on the warning information and sending it to the management personnel means predefining the warning information template through the Python database based on the issued warning information, inputting the warning information content into the warning information template, configuring the SMTP server, and sending the warning information template with the completed content input to the management personnel's email account via email, and sending the warning information template with the completed content input to the management personnel's mobile phone via SMS sending service.

[0178] Through communication technology, early warning information can be sent to relevant managers immediately after it is generated, improving the timeliness of early warning information transmission. By timely transmitting early warning information, it helps managers respond and take measures quickly, reduce the impact of natural disasters on agricultural production, provide real-time early warning information and decision-making support, and help emergency management departments take timely measures.

[0179] Example 2

[0180] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0181] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0182] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0183] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0184] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A natural disaster risk monitoring system based on satellite remote sensing images, characterized by: include, The data analysis module obtains environmental data from satellite remote sensing platforms and sensor platforms and comprehensively analyzes soil moisture; The drought risk prediction module monitors water demand in agricultural areas, analyzes water resource availability, and constructs a long-short-term memory network model to predict and monitor drought risks in agricultural areas. The artificial rainfall suitability analysis module uses a satellite remote sensing platform to identify cloud areas, analyze cloud parameters, identify clouds suitable for artificial rainfall, build a suitability prediction model to predict the suitability of clouds for artificial rainfall, analyze the spatial overlap of clouds and monitored agricultural areas, and determine the feasibility of artificial rainfall; The early warning information sending module issues early warning information based on water resource availability, drought risk and artificial rainfall feasibility, and generates templates based on the early warning information and sends them to management personnel.

2. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 1, characterized in that: The environmental data is obtained from the satellite remote sensing platform and the sensor platform, and the soil moisture is comprehensively analyzed, including: Acquire multispectral remote sensing image data for monitoring agricultural areas based on satellite remote sensing platforms, including visible light, infrared, and microwave data; Preprocess multispectral remote sensing image data, including radiation correction, atmospheric correction and geometric correction; Based on the satellite remote sensing platform, the reflectance of the near-infrared band, red light band and green light band is obtained, and the vegetation index NDVI and water body index NDWI are calculated; Based on the sensor platform, the soil moisture data of different layers in the agricultural area are collected and analyzed comprehensively, which is expressed as: Among them SM a represents the integrated soil moisture, SM i represents the moisture content of the i-th soil layer, and n represents the total number of layers.

3. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 2, characterized in that: The method is based on monitoring water demand in agricultural areas and analyzing water availability, including: Based on the sensor platform, available water resource data of monitored agricultural areas is obtained, including river water resource data, lake water resource data, and reservoir water resource data, which can be expressed as: WR r =WL·F; WR l =LL·A; WR t =RL·V; Among them WR r Indicates the water availability of the river, WR l Represents the water availability of the lake, WR t represents the water availability of the reservoir, WL represents the river level, LL represents the lake level, RL represents the reservoir level, F represents the water flow, A represents the lake area, and V represents the reservoir capacity; The comprehensive water availability is analyzed based on the available water resources in the monitored agricultural areas and expressed as: WR a =WR r +WR l +WR t ; Among them WR a represents the comprehensive water resources availability; Based on the historical data of available water resources, the mean and standard deviation of the historical comprehensive water resources availability data are calculated, and the comprehensive water resources availability data are standardized to calculate WR' a ; Taking into account the historical comprehensive water resources availability data and agricultural water use data, a historical threshold is set, which is expressed as: Where T represents the historical threshold, m represents the total number of historical time periods, represents the comprehensive water resource availability in time period i, AW i represents the agricultural water consumption in the i-th period, AW w represents the historical average agricultural water use; The threshold is adjusted according to the crop water demand in the currently monitored agricultural area and expressed as: Where T a Indicates real-time threshold, AW c Indicates the amount of agricultural water used in the current time period; If WR' a Greater than T a , it means that the available water resources in the monitored agricultural areas can meet the agricultural water demand; If WR' a Less than or equal to T a , it means that the available water resources in the monitored agricultural area cannot meet the agricultural water demand.

4. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 3, characterized in that: The long-term short-term memory network model is constructed to predict and monitor the drought risk in agricultural areas. include, Comprehensive water resources availability data WR' based on standardized processing a , construct a long short-term memory network model to predict and monitor soil moisture in agricultural areas, including input layer, LSTM layer, fully connected layer and output layer; The standardized comprehensive water resources availability data WR' a , vegetation index NDVI, water body index NDWI, comprehensive soil moisture SM a The land surface temperature data collected by sensors monitoring agricultural areas are time-aligned and fed into the model through the input layer; Use the training set to train the long short-term memory network model using the mean squared error loss function; Newly collected data are fed into the model to predict soil moisture in monitored agricultural areas; Based on historical data, the mean value of soil moisture minus the standard deviation is used as the high-risk threshold, and the mean value of soil moisture is used as the medium-risk threshold; If the predicted value of soil moisture y i If the value is greater than or equal to the high-risk threshold, the monitored agricultural area is judged to be at high risk of drought; If the predicted value of soil moisture y i If the drought risk is greater than or equal to the medium risk threshold and less than the high risk threshold, the monitored agricultural area is judged to be at medium drought risk; If the predicted value of soil moisture y i If the drought risk is less than the medium risk threshold, the monitored agricultural area is judged to be at low risk of drought.

5. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 4, characterized in that: The identification of cloud areas through satellite remote sensing platforms includes: Based on the multispectral remote sensing image data after radiation correction and atmospheric correction, the identification bands are selected for analysis. The identification bands include visible light band, near infrared band and infrared band. Set corresponding thresholds for visible light band, near infrared band and infrared band based on historical data; The identified band data that is greater than the corresponding threshold is binarized to segment and identify the cloud area.

6. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 5, characterized in that: The analysis of cloud parameters and identification of clouds suitable for artificial rainfall includes: Based on the satellite remote sensing platform, infrared band data is used to obtain cloud top temperature, and the cloud top height is calculated by combining the atmospheric temperature profile data, which is expressed as: Where CTH is the cloud top height, H0 is the tropopause height, T0 is the tropopause temperature, LR is the temperature lapse rate, and CTT is the cloud top temperature; The cloud base height is calculated using the wet bulb temperature method and is expressed as: Where CBH is the cloud base height, T is the surface temperature, and T d Indicates dew point temperature, F indicates temperature lapse rate; Based on the satellite remote sensing platform, microwave remote sensing data is used to obtain the water vapor density of each altitude layer in the cloud layer, and the cloud water path is calculated, which is expressed as: Where CWP represents cloud water path, p c (z) represents the water vapor density at the cloud height z; Based on the artificial rainfall operation conditions, the cloud water path, cloud thickness and cloud top temperature thresholds are set respectively, and the clouds that meet the threshold judgment conditions are marked as clouds suitable for artificial rainfall.

7. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 6, characterized in that: The method of constructing a suitability prediction model to predict the suitability of clouds for artificial rainfall includes: A suitability prediction model is constructed to judge the suitability of artificial rainfall based on the artificial rainfall operation conditions, which is expressed as: TR=w1·CWP+w2·CT+w3·CTT; Where TR represents the suitability of artificial rainfall, CT is the difference between the cloud top height and the cloud base height, and w1, w2 and w3 are weight parameters; Use the training set to train the model by minimizing the objective function, expressed as: Where c represents the total number of samples, r i Indicates the actual value, Represents the model output; The model parameters are iteratively optimized using the gradient descent method. When the model loss no longer decreases significantly during the continuous iteration process, the iteration is stopped and the model parameters are output to update the fitness prediction model. The suitability of artificial rainfall was calculated using the suitability prediction model based on historical sample data, and a high suitability threshold T was set based on the 90th percentile. a , set the suitability threshold T based on the sample median z ; If the artificial rainfall suitability TR is greater than or equal to the high suitability threshold, the monitored cloud layer is determined to be suitable for artificial rainfall; If the artificial rainfall suitability TR is greater than or equal to the medium suitability threshold and less than the high suitability threshold, it is determined that artificial rainfall can be carried out in the monitored cloud layer; If the artificial rainfall suitability TR is less than the medium suitability threshold, the monitored cloud layer is judged to be unsuitable for artificial rainfall.

8. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 7, characterized in that: The analysis of cloud layer and spatial superposition of monitoring agricultural areas to determine the feasibility of artificial rainfall includes: Obtain cloud position coordinates from multispectral remote sensing image data based on the identified cloud area; Based on the GIS boundary data of the monitored agricultural area, the cloud position coordinates are mapped into the GIS system and displayed on the same map with the GIS boundary data of the monitored agricultural area. Spatial overlay analysis is performed, and the spatial overlay with the cloud data is judged according to the monitored agricultural area, including inclusion judgment and intersection judgment; The median of the intersection judgment is set as the boundary threshold. If the intersection judgment value of the cloud data and the monitored agricultural area is greater than the boundary threshold, and the intersection judgment of the cloud data and the monitored agricultural area is included, the analysis result is suitable for artificial rainfall. If the intersection of cloud data and the monitored agricultural area is less than or equal to the boundary threshold, the analysis result is that it is not suitable for artificial rainfall.

9. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 8, characterized in that: The warning information is issued based on water availability, drought risk and feasibility of artificial rainfall. include, Based on the comprehensive water resource availability and real-time threshold determination of the monitored agricultural area, an agricultural water shortage warning message is issued when the available water resources in the monitored area cannot meet the agricultural water demand; Based on the predicted soil moisture values ​​of monitored agricultural areas and the determination of high and low risk thresholds, drought warning information will be issued when it is determined to be a moderate or high drought risk; Based on the cloud areas identified in the monitoring of agricultural areas, and the determination of the suitability of the cloud layers for artificial rainfall and the high suitability threshold and the medium suitability threshold, when it is determined that artificial rainfall is not suitable, an early warning message that artificial rainfall is not suitable is issued.

10. The natural disaster risk monitoring system based on satellite remote sensing images according to claim 9, characterized in that: The generating template according to the warning information and sending it to the management personnel refers to predefining the warning information template through the Python database based on the issued warning information, inputting the warning information content into the warning information template, configuring the SMTP server, sending the warning information template with the completed content input via email to the management personnel's email account, and sending the warning information template with the completed content input via SMS sending service to the management personnel's mobile phone via SMS.

Images